PROJECT SUMMARY Current treatment of hemophilia A (HemA) patients with repeated infusions of factor VIII (FVIII; abbreviated as F8 in constructs) is costly, inconvenient, and incompletely effective. In addition, ~25% of treated patients develop anti-FVIII immune responses (inhibitors); it is particularly challenging to treat these inhibitor patients. Gene therapy that can achieve long-term phenotypic correction without the complication of anti-FVIII antibody formation is highly desired for patients with or without inhibitors. Recently it was shown that FVIII expressed ectopically in megakaryocyte (Meg) is effective to prevent HemA mouse tail bleeding to death. Ex vivo HSC gene therapy corrected the bleeding diathesis even in the presence of inhibitors. This is because FVIII synthesized in Megs is stored in ?-granules and only released at the injury sites during platelet activation, therefore protected from circulating inhibitors. However, several limitations exist for ex vivo gene therapy. First, FVIII stored in platelets (pFVIII) has different temporal-spatial availability compared with plasma FVIII. pFVIII was shown to have variable efficacy in different hemostasis mouse models and its resistance to inhibitors also varies in different settings. The functional roles of gene therapy delivered pFVIII and its resistance to inhibitors will need to be carefully investigated and defined. Second, there are difficulties encountered by ex vivo HSC gene therapy, in particular, preconditioning required by ex vivo gene therapy is highly undesirable for HemA patients. Third, it was shown that high levels of pFVIII expression can induce platelet apoptosis. Preconditioning regimens used by ex vivo gene therapy induce thrombocytopenia. This problem can be compounded with pFVIII expression, leading to significant thrombocytopenia, which poses a major hemostatic risk to HemA inhibitor patients. Recently, intraosseous (IO) delivery of lentiviral vector (LV) has been shown to effectively transduce bone marrow (BM) cells in mice. We propose to employ this new approach to deliver F8- LVs into BM, which avoids the use of preconditioning regimen, thereby decreasing the associated risk with thrombocytopenia and can be both safe and efficacious for clinical use. Previously, we demonstrated a single IO delivery of F8-LVs driven by a human Meg-specific glycoprotein 1b? (GP1b?) promoter (G-F8-LV) produced platelet-specific FVIII expression, leading to long-term, partial correction of HemA mice both with and without pre-existing inhibitors. In the current proposal, we will expand our studies targeting FVIII expression in platelets via IO delivery, aiming at optimizing this novel technology and developing preclinical protocols in humanized mice and large animal models (HemA dogs). AIM 1. Optimize IO delivery of G-F8-LVs in mice and examine the resulting biological efficacy of pFVIII. AIM 2. Optimize pFVIII gene expression in human Megs. AIM 3. Evaluate if IO delivery of LVs can effectively transduce HSCs in large animal models and if IO delivery of canine F8 (cF8)-LV can correct the phenotype in HemA dogs.